VMD-L Mailing List

From: Dan Lussier (dtlussier_at_gmail.com)
Date: Tue Aug 24 2010 - 12:59:40 CDT

Ok - thanks - that clear it right up. I'll do some converting as
necessary from now on as you suggest.

Dan

On 24-Aug-10, at 9:48 AM, Axel Kohlmeyer wrote:

> hi dan,
>
> On Tue, Aug 24, 2010 at 10:38 AM, Dan Lussier <dtlussier_at_gmail.com>
> wrote:
>> Hi Axel,
>>
>> Ok - that is good to know. One off the reasons I asked at all was
>> because
>> when I load my LAMMPS dump I am observing a strange (or at least it
>> appears
>> strange to me) behaviour.
>>
>> In my case the x coordinate is limited from [0,160] but when the
>> data is
>> loaded into VMD and I look at the range of x I find something like
>> [0,25000]. The same thing is seen in both y and x so the overall
>> effect
>> appears to be applied evenly to all the coordinates such that the
>> geometry
>> is simply scaled rather than mangled.
>
> ok. that means, you have coordinates that somehow manage
> to fool the heuristics to tell scaled from unscaled coordinates
> for old style dump files. unusual, but not impossible. this is one
> of the reasons that prompted the change in the file format.
>
>> I am using unscaled coordinates if this is of consequence to the
>> way the
>> reader works.
>
> yes.
>
>> It usually isn't a big deal - but it makes interacting with the
>> data via
>> selections and scripting more tricky, so if there was a way to
>> avoid this
>> I'd love to hear.
>
> the best way would be to "convert" them to the new dump style.
> the fastest is probably to use the sed command. e.g. with:
>
> for s in *.lammpstrj ; do sed -e 's,^ITEM: ATOMS,ITEM: ATOMS id type x
> y z,' $s > tmp && tmp $s; done
>
> or something equivalent for the second ITEM string that matches your
> dump file.
>
> cheers,
> axel.
>
>> Thanks again,
>>
>> Dan
>>
>> On 23-Aug-10, at 8:04 PM, Axel Kohlmeyer wrote:
>>
>>> On Mon, 2010-08-23 at 19:08 -0500, Dan Lussier wrote:
>>>>
>>>> Hi,
>>>
>>> dan,
>>>
>>>> I have some old (i.e. not self-describing) LAMMPS ASCII dump files
>>>> that I would like to look at again in VMD.
>>>
>>>> Is it possible to simply add the necessary column definitions
>>>> (i.e. id
>>>> type x y z vx vy vz ..) to the 9th line of the dump file (i.e.
>>>> ITEM:
>>>> ATOMS) so that it can be read properly by the newest version of the
>>>> LAMMPS reader plugin?
>>>
>>> the current lammps plugin should still support old style dump files
>>> in the same way it was supporting them before. i've recently
>>> fixed some bugs and added support for triclinic cells, so you may
>>> want to try one of the latest VMD alpha test versions or download
>>> and
>>> compile the plugin from the sources posted here (only tested on
>>> linux):
>>> http://sites.google.com/site/akohlmey/software/lammps-icms
>>>
>>> if you want to also read in velocities, which was previously not
>>> supported, you would have to add the description list, but only
>>> for the first frame in each file.
>>>
>>> cheers,
>>> axel.
>>>
>>>> Thanks.
>>>
>>> --
>>> Dr. Axel Kohlmeyer akohlmey_at_gmail.com
>>> http://sites.google.com/site/akohlmey/
>>>
>>> Institute for Computational Molecular Science
>>> Temple University, Philadelphia PA, USA.
>>>
>>
>>
>
>
>
> --
> Dr. Axel Kohlmeyer akohlmey_at_gmail.com
> http://sites.google.com/site/akohlmey/
>
> Institute for Computational Molecular Science
> Temple University, Philadelphia PA, USA.